Provenance Research Methodologies
The purpose of the pilot project (January 2020-July 2021), The Fate of the Adolphe Schloss Collection, co-funded by the European Union, was to test the feasibility of creating an event-based database of historical losses of cultural objects during the Nazi era. To that end, the project selected a collection of 333 paintings assembled by Adolphe Schloss in Paris (France) between 1870 and 1910. During the German occupation of France (1940-1944), French and German officials confiscated and dispersed the Schloss Collection.
In order to build the database, detailed historical information was needed about:
the objects which comprised the collection;
the people and events surrounding the creation and evolution of the collection;
the events that led to its confiscation, dispersal, and recovery.
The data model incorporates events and people, and presents the objects comprising the collection. Those pieces of data are contained in extensive archival documents and art historical sources that the project compiled over a year starting in the spring of 2020.
As the research evolved, the data model grew more complex, reflecting the interweaving of facts, people, places, organizations and objects, across time and space. However, throughout the duration of the pilot project, archives and specialized libraries were mostly closed to the public due to the pandemic. Whenever possible, sporadic research was conducted throughout the summer and fall of 2020 so as to collect as many images of archival documents as possible, building a solid basis of historical data for extraction and analysis.
Objects
Object-based research—focused on the paintings and their creators—is now firmly established as part of provenance research.
Inventories
The paintings are described in inventories of the collection drawn up at different times (1923, 1936, 1943-1945) and for different reasons. Once the paintings comprising the Adolphe Schloss Collection were identified, information was gathered about the artists and about each work’s historical context, location and date of its creation, its size, and medium. This constitutes the starting point for reconstructing the history of an art object.
The principal inventories used in the pilot project were:
1923 inventory drawn up by Clothilde Brière-Misme, Bibliothèque d’Art et d’Histoire, Paris
1938 inventory attached to Lucie Haas Schloss’ will
1943 (May-June?) inventory drawn up by Jean-François Lefranc at the Banque de France, Limoges, after confiscation
1943 (August 13) inventory drawn up by French officials at the Banque Dreyfus, HQ of the Commissariat Général aux Questions Juives, where the Schloss Collection was unpacked and catalogued after confiscation, each painting given a number
1943 (August 20) inventory drawn up by Louvre officials after they exercised a right of pre-emption on 49 works from the Schloss Collection to be incorporated into the Dutch and Flemish collections at the Louvre
1943 (November 3) inventory drawn up by German officials cataloguing the paintings from the Schloss Collections and assigning numbers 1-280 to the remaining paintings
1944 (date uncertain) two German inventories drawn up most likely at the Führerbau in Munich
1945 (February 15) incomplete American inventory
In order to reconstruct the complete list of works collected by Adolphe Schloss, it was necessary to compare these inventories. Using the provenance information contained therein for many of the Schloss paintings, the search was broadened to include:
exhibition records
catalogues and catalogues raisonnés
monographs
newspaper articles mentioning the item
exhibition histories
bibliographies of the artist and the movement to which he/she belonged
business/personal correspondence between owners and various officials
wills and estate settlements
The project amassed photographs of the paintings taken at key stages, to be used as illustrations. The originals are located in archives and specialized libraries in the United States (New York) and Europe (Paris, The Hague, Koblenz, Munich). Like the inventories, these photos date back to the mid-1920s to 1943. Photographs also exist of the interior of the Schloss residence at 38 Avenue Henri Martin in Paris, whose walls were covered from floor to ceiling with the paintings collected by Adolphe Schloss. These photos are found at the Paris-based Musée d’Art et d’Histoire du Judaïsme (MAHJ).
While reconstructing the Schloss Collection, dozens of paintings came to light that Adolphe Schloss had bought and sold over the years. They were recorded in the database as an extension of Schloss’ collecting efforts and expressions of his tastes and preferences.
Photographs
The apparent first set of photographs was produced in 1925 at the request of Helen Clay Frick of the Frick Art Gallery. Clothilde Brière-Misme of the Bibliothèque des Arts et de l’Histoire complied with Frick’s request, authoring the first known inventory of the Schloss Collection, dated 1923. It is not clear whether the Frick museum received photos of all the paintings or of a selection thereof.
The next time the Schloss Collection was photographed was after its confiscation and transfer to the Banque Dreyfus in Paris in August 1943. A photographer was hired for the Linz Museum project. Erhard Göpel, representing the Sonderauftrag Linz, paid for the set of photos produced for the Linz Museum project. The photographer’s name was Mr. Gauthier. Did the Louvre hire a photographer?
Bruno Lohse and Göpel held a complete set of photographs of the paintings in the Schloss Collection. Göpel held the negatives while Lohse assembled a two-volume set of prints of the Schloss paintings that he had bound in red leather to offer to Göring on his and Göpel’s behalf.
The photos in the Ministry of Foreign Affairs archive may come from the Frick, the Louvre, and Linz sets.
Not all the datasets for recovered Schloss paintings featured in the Munich Central Collecting Point database carry photos with them. Some of the photos appear to be of a lesser quality than those in the Koblenz files.
Two curious items are the photo albums of the Schloss Collection in the Koblenz archives under B323/189 and B323/190. They include photos of the Linz Museum paintings and those that went to the Louvre. B323/189 is entitled: “Schloss Collection: retrieved objects,” an indication that perhaps this album was either produced at the Munich Central Collecting Point or that the Americans found it and annotated it when they recovered paintings from the Schloss Collection. The photo album in B323/190 is devoted exclusively to the Schloss paintings pre-empted by the Louvre museum. The labels are typed in French, thus leading one to believe that the Louvre may have produced this album. Interestingly enough, the labels on these photos are the same as those that appear on the backs of some photos in the Ministry of Foreign Affairs archive at La Courneuve, north of Paris.
Markings
Sometimes, the backs of paintings are covered with markings—labels, stamps, inscriptions, numbers, labels and stickers, and descriptive text about the work. Since it was not possible to visit the museums that currently hold paintings from the Schloss Collection, it was necessary to rely on archival documents on which the markings were recorded and on the assistance of museum professionals to supply photos of the backs of their paintings from the Schloss Collection in order to enrich the understanding of the paintings’ journeys.
These markings include, but are not limited to:
stamps from dealers, museums, private collections
inscriptions from the artist to the buyer or a dedication to an individual
numbers from an inventory, an auction house sale, a storage or warehouse
labels from shipping companies, customs agencies, other government entities, galleries, exhibitions, which might include the title and author of the work, measurements, inventory numbers.
The most important source of information on the markings found on the backs of Schloss paintings were the registration cards at the Munich Central Collecting Point (MCCP), established by the U.S. Army in mid-May 1945 as a central processing center for all looted art objects found in the U.S. zone of occupation of Germany. The cards were filled out for each object brought to the MCCP; one of the boxes to fill out on the card was labeled “identifying marks,” for information collected from the back of the recovered object.
80 MCCP cards for Schloss paintings captured one or more markings.
32 paintings still had the original crate label and crate number dating back to August 1939.
55 paintings carried the French ID number extracted from the so-called Banque Dreyfus inventory of August 1943.
66 paintings carried the German ID number extracted from the November 1943 German inventory of the Schloss paintings earmarked for the Linz Museum project.
The principal markings included:
A label indicating the number of the crate, an inventory number, and the artist’s name: Caisse 6, No. 133, M. Koffermans
Exhibition labels
Provenance Research Challenges: Gaps and Uncertainties
One of the challenges underlying the research into the history of ownership of an art object is the question of how to address gaps in knowledge about its history and uncertainties surrounding the available historical information. The ultimate goal of research is to reduce the size and scope of any gaps.
Representing gaps and uncertainties in a data model
Information gaps in the history of ownership of an object are as ubiquitous as the uncertainties surrounding individuals and entities associated with its history.
Gaps and uncertainties are treated as “fuzzy data,” with unclear start and end points, no well-defined temporal (dates) and spatial (locations, people, entities) edges. While qualifiers are attached to uncertain information in a provenance (likely, perhaps, probably, maybe, circa, etc.), no such conventions exist for representing a gap or lacuna in the historical data. For now, the pilot project uses qualifiers to address uncertainties and tags a provenance gap with “unknown owner(s).”
What is a gap in the provenance of an object?
A provenance gap is a temporal and spatial disruption in the accounting for an object’s whereabouts. Our challenge is to mobilize intellectual and technological resources to “narrow” the gaps with contextual material that could lead to research breakthroughs. The gap is not empty. It represents content that lies behind a door, blocked from our view by an obstruction. The key is to eliminate the obstacle, the obstruction, and gain access to the content. But how?
Edges constitute the contours of a gap in time and space. Edges are markers for place, date, person, entity and event, or any combination thereof. External or contextual events in and through which objects move can reduce the size and depth of a gap. By exploring the edges of the gap, one can build bridges of circumstantial evidence over it, creating a research context on how to narrow its breadth.
What is uncertainty in historical data?
Uncertainties in data occur when an exact date is not known, an individual or entity associated with the history of an object cannot be verified through independent research. When there are several purported owners sharing the same last name and living in the same period of time, it is often impossible to tease out which one is the correct owner or possessor of the object in question.
Gaps and uncertainties in the recorded history of Schloss 260
[Note: uncertainties are in italics; gaps are in bold]
Schloss 260
Adriaen BrouwerPossibly painted ca. 1630 (uncertainty)
Ca. 1630-1910 (gap)
When did the painting leave the artist’s studio?
Did he sell it before 1638? [Brouwer died in January 1638, Antwerp] If not, who inherited it?
Where did it go? Did it remain in Antwerp? Until when? With whom? Where?
When did the painting cross into France?
Did the painting enter France from Antwerp or some other location?
When, where and from whom did Adolphe Schloss buy the painting?
Private collection Adolphe Schloss (1842-1910), Paris until 1910
…..
Ca. November 1943: Jean-François Lefranc, Paris (uncertain date and place)
Cornelis Postma, Paris (uncertain date and place)
Sold to Henri Verne for 300,000 francs, late 1943/early 1944 (uncertain date and place)
Sold to Etienne Nicolas, Paris, for 330,000 francs, early 1944 (uncertain date and place)
1944-June 1947 (gap)
Was Nicolas the only person to possess the work from 1944 to 1947?
When and where was it recovered? By whom?
By 11 June 1947: Restituted to Schloss heirs (uncertain date)
…
1951-2015 (gap)
Who bought Schloss 260 at the Galerie Charpentier sale?
Did the painting remain in Paris? In France?
Did it cross borders? Where? When? With whom?
Was it exhibited? Where? When? By whom?
How many times did it change hands before its 2015 sale in Brussels?
Sold at Vanderkindere, Brussels, Belgium, 13 January 2015 as Adriaen van Ostade
Transforming the gap into a set of uncertainties
The purpose of solving the gap problem is to reduce its size and scope so that, from a data standpoint, it turns into a set of certain and uncertain pieces of historical information, much like filling in a jigsaw puzzle. One way to reach that goal is through multi-pronged research and logical reasoning. In the case of Schloss 260—a painting by Adriaen Brouwer—the research should focus on Adriaen Brouwer and the circulation of his works in his lifetime and years following his death. Another focus point is to backtrack from Adolphe Schloss’ ownership and elicit the name(s) of the person(s) from whom he acquired the Brouwer painting, where and when. Assuming that he bought the painting in France, a comparative analysis of other Dutch and Flemish works from the Adolphe Schloss Collection leads to these possible paths:
from the Netherlands to the British art market and then to France;
directly to France, then to the UK and back to France;
directly to France from the Netherlands;
through another country (Belgium, Germany, are the most likely candidates) and then to France.
A systematic survey of art sales and exhibits involving works by Brouwer would focus on art dealers who regularly supplied Adolphe Schloss with old master paintings in the late nineteenth century and early twentieth century. A thorough review of the Brouwer literature—exhibition catalogues, scholarly monographs, catalogues raisonnés—would also be warranted.
Reducing uncertainties
The methodologies used for reducing the size and scope of gaps apply as well to resolving uncertainties surrounding historical data in the provenance of an object. With regard to Schloss 260, the most important uncertainties pertain to the trajectory of the painting from the time that Jean-François Lefranc gained custody of it (how and where did the transfer actually take place?) and its subsequent sale to various individuals in Paris. The date and location of the transfer of the Brouwer painting from Lefranc to Postma to Verne to Nicolas is not yet known. The answer most likely resides in historical documents in Paris-based archives. The few documents that were collected confirm that the above-named individuals were in direct contact with Schloss 260 at some point between late 1943 and 1944 but without specifying time and place. Equally important is the absence of information on how and where the French authorities recovered the painting.